in the Yuma project; 415,000 acres in the Imperial Valley; and 190,000 acres south of the international boundary line-a total of 698,000 acres. This total is almost exactly double the acreage irrigated in 1913, showing the rapid rate of increase in the use of water in the lower basin. The possible extension of irrigation in the lower basin has not been determined fully, but conservative estimates indicate that the following additional areas can be brought under irrigation: 260,000 acres between Needles and Yuma, 150,000 acres of which is on the Arizona side; 76,000 acres in the Yuma project; 400,000 acres in the Imperial and Coachella Valleys; and 630,000 acres in Mexico.

WATER SUPPLY.

Engineers have methods, of comparative accuracy, for measuring the quantity of water flowing in rivers. The record of the flow, day by day, month by month, shows the extent of the water supply and its fluctuations, and furnishes a basis for the design of engineering works. On the Colorado River and its tributaries, many gauging stations, at carefully chosen locations, have been kept for varying periods of time, some of the records extending over 25 years.

The records of stream flow at Yuma have been kept since January, 1902. The gaging station is below the mouth of the Gila River and below the Yuma diversion dam, but above the head gates of the Imperial Canal. The average annual flow at the gaging station for the period 1902-1920 was 17,300,000 acre-feet. Had the present irrigated area been under irrigation throughout the period of the record, the average annual flow would have been about 16,000,000 acre-feet. The average flow at Boulder Canyon is practically the same amount, since diversions and losses between Boulder Canyon and Yuma are balanced by the inflow of tributaries.

Most of the water comes from the upper basin. At the junction of the Grand and the Green, the average annual discharge of the Grand is 6,900,000 acre-feet, and of the Green 5,500,000 acre-feet. The Green and Grand and San Juan Rivers together, though draining less than two-fifths of the area of the Colorado Basin, furnish 86 per cent of the total water supply.

By far the greater part of the precipitation in Colorado and Wyoming is in the form of snow. During the winter the snow accumulates to great depths. The melting of the snow during the spring months produces a long period of high water, the annual flood, which lasts from two to three months and reaches its highest point at Yuma usually in June. During the June flood of 1909, the flow at Yuma reached 150,000 cubic feet per second. On June 27, 1921, all previous June records were broken by a flow of 186,000 cubic feet per second. The low-water season begins in August and lasts from three to seven months. The minimum flow at Yuma has been below 4,000 cubic feet per second during several low-water seasons.

The Gila River drains an area of 57,000 square miles. While the average annual discharge of the river is not great it is very variable. In 1916 the discharge of the river at its mouth was 4,500,000 acre-feet; in some other years the total has been less than 100,000 acre-feet. Short-lived, "flashy" floods, greater than the highest peak floods in the Colorado, occur at times. The flow on January 16, 1916, reached 220,000 cubic feet per second. It is fortunate that the Gila floods do not come at the same time as the Colorado floods, in May or June. Should they coincide, the menace to the Yuma and Imperial valleys would be intensified; the levees would be overwhelmed.

RESERVOIR SITES.

There are scores, hundreds, of storage sites in the middle and upper parts of the Colorado Basin. Many of them have been surveyed, and at several of the sites the depth to bedrock has been ascertained by diamond drilling. The Strawberry Valley site in Utah and the Roosevelt site in Arizona and some small sites have been occupied already. For the complete regulation and utilization of the river, there are adequate natural opportunities: the real problem is as to which is the best. A few of the largest and most promising sites, those which are of greatest public interest, will be discussed.

The Dewey Reservoir site is situated on the Grand River just west of the UtahColorado line. Although one of the last to be discovered, it is one of the best. It is the only site for a large reservoir on the Grand River except the Kremmling, and that site is occupied by a railroad. The bedrock at the Dewey site is only 44 feet. below the river bed, and the capacity with a dam only 215 feet high from river bed to spillway is 2,300,000 acre-feet.

The Flaming Gorge site is on the Green River in Utah just south of the Wyoming line. The greatest depth to bedrock is 73 feet, and a 215-foot dam will impound 3,120,000 acre-feet. The width of the canyon is 200 feet. The Flaming Gorge and the Dewey sites control the most important headwaters of the Colorado. Both are excellent projects and should be under construction to-day.

Another excellent site is on the Yampa tributary, near Juniper Mountain. The drainage area is small, but the stream flow approximates 1,000,000 acre-feet of water per year. A 200-foot dam would provide a capacity of 1,500,000 acre-feet. The depth to bedrock is 24 feet.

The Ouray Reservoir site is on the Green River a hundred miles below the Flaming Gorge site. This site is remarkable in that a dam only 210 feet high would impound 16,000,000 acre-feet of water. The greatest depth to bedrock, a factor of great influence on the cost of a dam, is 121 feet, and the canyon is not narrow. This site should be held available by the Federal Government until it is absolutely certain that the site is not needed in the general scheme for development of the river. If the site is restored to entry, it will be seized at once by the Denver & Salt Lake Railroad. The railway can be built around the reservoir site.

A reservoir at the junction of the Green and the Grand has been under consideration for many years. It would regulate partially both streams. A dam 250 feet high would impound 7,450,000 acre-feet. Borings were made to 120 feet without encountering bedrock. It is unfortunate that the borings were not carried somewhat deeper. An apparently excellent reservoir site exists on the San Juan near Bluff, Utah, but its feasibility has not been established by test holes to find bedrock. A dam 264 feet high would create a reservoir of 2,600,000 acre-feet capacity. The accumulation of silt in this reservoir would be very rapid.

The Glen Canyon, or Lees Ferry, site outclasses all other proposed sites in its gigantic possibilities. The maximum development contemplates a dam 700 feet high, 450 feet long at the level of the river, and 1,400 feet long on top. The proposed slopes are one to six downstream and one to four upstream, making the length of base up and downstream over a mile. The capacity of the reservoir would be over 50,000,000 acre-feet, and 86 per cent of the entire water supply of the Colorado Basin would be regulated completely. Over a million continuous horsepower could be developed without sacrifice of irrigation interests. Complete surveys of the reservoir site have been made during the last few months. No test borings have been made, and it is stated that the depth to bedrock is not a crucial matter on account of the radial character of the dam contemplated. Test borings should be made at once.

Excellent dam sites exist in Cataract Canyon and Marble Canyon. The project for Marble Canyon provides for a power development of 1,300,000 horsepower, but the storage possibilities are small. This will be the last of the major projects because of its magnitude and high cost.

On the Little Colorado River there is a dam site at Tolchaco, where the entire flood flow of that stream can be controlled by a dam 50 feet high.

The site at the mouth of Diamond Creek is of particular interest to Arizona, on account of its favorable location and because it is controlled by Arizona people. The site is only 16 miles from Peach Springs, a station on the Santa Fe Railroad. It is a power project only, there being practically no storage. Present plans, subject to modification, call for a dam 284 feet high, 324 feet above bedrock, to the spillway crest, and the top of the structure would be 25 feet higher. About 110,000 horsepower could be developed with the unregulated flow of the river, but in case the flow is equalized by a project with storage farther up the river the ultimate power development may reach 600,000 horsepower. The canyon at this site is only 220 feet wide at the water level, and the length of the dam at the top will be 600 feet, about the same as the Roosevelt Dam. The walls and foundation are of granite. The main electric transmission line would extend through, or near, Prescott, Phoenix, Mesa, Florence, and Tucson to Douglas, with important laterals to Jerome, Ray, Globe, Clifton, Ajo, and Yuma.

The Boulder Canyon site is in a similar narrow canyon in granite rock. The canyon walls are 300 feet apart. Here it is proposed to build a solid concrete masonry dam 600 feet high, 735 feet above bedrock, to elevation 1,300 feet above sea level. The capacity of the reservoir is 31,600,000 acre-feet, and the estimated cost of the dam alone is $55,000,000. The great depth to bedrock is the main disadvantage of this site. While the problems of carrying the foundation to so great a depth and of passing the annual and occasional floods of the river during the construction period strike terror to the heart of the engineer, the task can be accomplished if adequate funds are provided. The power development will be 700,000 continuous horsepower as long as the irrigated area in the lower basin does not exceed 1,500,000 acres, and will decrease to 600,000 horsepower as the acreage increases to 2,000,000 acres.

The last annual report of the United States Reclamation Service states that an inspection of the lower river was made by boat by Homer Hamlin, a noted engineer, in April, 1920, and that he reports that there is no good dem site for a storage reservoir between Boulder Canyon and Yuma.

THE THREE GREAT PROBLEMS.

Three objects are sought in the development of the Colorado River. They are: (1) Storage for flood protection; (2) storage to provide more water for the latter half of the irrigation season and for dry years; and (3) hydroelectric power.

The flood protection is the main incentive which is spurring many agencies to action. The people of the Imperial Valley, for 16 years, have been fighting a defensive battle against the Colorado, sometimes gaining, sometimes losing, but in the main losing. They can not hold out for many more years. At least once every year, in June, and sometimes at other seasons, the river threatens to change its course from the Gulf of California to the Imperial Valley, as it did in 1905. The only protection at present is the system of levees, called respectively the first, second, and third lines of defense. Frequently the floods break through the first and second lines and reach the third line. Each year the river, through silt deposition, builds up that part of the alluvial fan in front of the levees, in some years as much as 4 feet, and each year the levees must be raised an equal amount. Over one-quarter of a million dollars is expended each year by the farmers of the Imperial Valley in this work. The limit will be reached soon. Levees 40 or 50 feet high can not be maintained.

The Yuma Valley also is protected by levees, but the danger there does not increase. Arizona hopes to develop another great irrigated valley farther upstream at Parker, but much of the Parker Valley is now subject to overflow and must be protected by an expensive system of levees unless adequate regulation of the flood waters is provided by storage reservoirs. Regulation of the Green and the Grand will solve the problem in large measure, but tributaries below the junction must be given consideration. On one occasion a flood of 150,000 second-feet measured at Bluff, Utah, was contributed by the San Juan, and the Gila River floods likewise are a menace with which to reckon. As for storage to equalize the supply for irrigation, the situation is more critical than is commonly known. Despite the great excess of water which is wasted to the ocean each year, there is an actual shortage during the latter part of the irrigation season in dry years. In 1915 the entire flow of the river was diverted into the Imperial Canal at the end of August, and yet there was not enough water to meet the demand. Since that time the acreage irrigated from the river has increased 300,000 acres. If the natural flow next September is as low as it was in 1915, there will be 300,000 acres of crops without any water to bring them to maturity, and the financial loss and human suffering will be appalling. Again, it is the Imperial Valley that is in danger, for other projects have the advantage of location upstream. No further expansion of irrigation use should be allowed until storage is provided; it should be admitted that the natural flow is entirely appropriated. It does not seem practicable, however, to prevent continued appropriation and use of water in Utah and Colorado.

But how can storage be financed? The Imperial Valley is burdened already with a heavy bonded indebtedness and is facing the further problem of the all-American canal, which is expected to cost $30,000,000. The farmers can not finance the river regulation which they require and must have.

Now enters the third element of the great project-power. The power possibilities are so great, and power is so valuable, that it is estimated the sale of power will pay for the entire project. A few months ago the proposal was to charge 5 per cent of the cost of the storage dam to irrigation, 10 per cent to flood protection, and 85 per cent to power. Now it is proposed to charge the entire cost to the power privileges. About 4,000,000 horsepower can be developed in Arizona at the four sites mentioned above. Is there a market for so much power? Arizona can take about 100,000 horsepower to replace present steam plants. Cheap power will permit of increased pump irrigation, the mining of lower grade ores, and the electrification of our railways. We shall have factories where our own raw materials can be fabricated, cotton mills, copper and brass foundries; and the electrolytic refining of Arizona copper can be done in our own State. All city and house lighting will be done with hydroelectric power, and any excess can be used for making nitrate fertilizers.

But other States, especially California, will compete for the power. A great amount can be marketed in southern California now. It is estimated that in 15 years all possible hydroelectric development in that State will have been accomplished, and California interests are looking much farther ahead than that.

Nearly all of the power requirements of the mining industry in Arizona are now supplied from petroleum fuel oil. The best opinions regarding the future supply of fuel oil point to a diminution of the supply and rapidly rising prices. It is essential that hydroelectric power be developed to replace the failing oil supply.

PROPOSAL OF THE UNITED STATES RECLAMATION SERVICE.

Engineers of the United States Reclamation Service have been studying the problem of the Colorado for eight years, and have decided quite definitely on what they believe should be the first project. The service has recommended to Congress that it should be a project of the Federal Government, and the Secretary of the Interior stated publicly at the Riverside and San Diego conventions in December, 1921, that, because of the international and interstate character of the river, the Federal Government is the only competent agency to construct the great dam that must be built, and to control and operate its gates. He is right, and Arizona should back to the the limit Federal ownership and operation of the main river-control project.

The Reclamation Service recommends that the dam be located in Boulder Canyon, on the boundary line between Arizona and Nevada. On account of the peculiar situation, the west end of the dam would be rest on the Arizona side. A transmission line from that point to Phoenix would be about 250 miles long, and a line to Los Angeles 277 miles in length. The proposed 600-foot dam provides for storage for irrigation and for storage of silt for 60 years, and for 5,000,000 acre-feet capacity at the top to be used only for detention of high-flood crests, such as those of 1907, 1909, 1914, and 1920.

Last July, when Congress was committed to retrenchment, and it seemed impossible to interest the East in this most necessary undertaking, plans were made to contract the power privileges in advance to municipalities and States or to other purchasers, and the purchasers were to obtain the necessary funds through sale of bond issues. The city of Los Angeles was ready to take all or as much of the power as would be allowed to that city. Now, it is believed that there is a good fighting chance to obtain the money through Federal appropriation, with ultimate return of the cost to the Government by the sale of power.

ALTERNATIVE PROPOSALS.

Although crystallization of sentiment in favor of Boulder Canyon project has made considerable headway, still some widely divergent views are being expressed, and it may not be impertinent to discuss alternative proposals. It is contended that for many reasons the river development should begin farther upstream. That the Boulder Canyon site is the one nearest to the best market for power is a sound argument. Of the other arguments advanced for that site, some are not valid, and the others may be met by the statement that extensive storage in the upper basin can be followed advantageously, and will be, by projects providing additional storage on the lower river. If the flood hazard is removed or is greatly reduced by means of extensive storage in Utah, the Boulder Canyon Dam can be built at much less cost and in fewer years. Further, if the river regulation is effected in the upper basin, the power sites from Glen Canyon to Boulder Canyon, inclusive, become much more valuable, since the water supply is equalized and because less reserve space is required for detention purposes. The upper locations will be developed eventually; why not now?

From that standpoint the Dewey site on the Grand River and the Flaming Gorge site on the Green offer the best solution. Both dams could be built at once, and the total cost would be only about $25,000,000. The Juniper Mountain Reservoir would cost $4,000,000. These sites are above the great silt-gathering area of the drainage basin. The Flaming Gorge and Dewey Reservoirs would provide ample late-summer water supply for the lower basin for many years to come. The Flaming Gorge Reservoir would serve to reduce the spring floods on the Green River one-third, and the Dewey Reservoir would take the peak off from the spring floods of the Grand. The Dewey Reservoir would be operated so as to be entirely empty at the beginning of the flood period. Both dams could be completed in five years. It is premised, however, that the construction of these dams would be followed by that of one or more others farther downstream-possibly one on the San Juan or at Lees Ferry, and either the Diamond Creek Dam or Boulder Canyon Dam or both. The dams on the headwaters should be built under the same theory of government as were the 33 dams on the Ohio River, that is, to secure river regulation and control, to make the stream manageable and utilizable. Navigation is no more vital to the economic and social welfare of the group of six States bordering the Ohio than is the taming and harnessing of the Colorado to the welfare of the seven States along its course. In due time the Government might be reimbursed for the investment, for, after the construction of large storage reservoirs in Arizona, the Utah reservoirs would be of great value for power production.

The Diamond Creek project is capable of comparatively rapid construction, and is quite likely to go ahead of the Boulder Dam in point of time. It would be a strictly

Arizona enterprise, and free from the entangling jurisdictions that are inevitable in the larger projects. It does not in any way lessen the necessity for the Boulder Dam or some other dam which can provide storage and flood control.

Another proposal is to make Lees Ferry Reservoir the first major undertaking. On account of the type of dam planned, the extent of flooding in the river during construction would be immaterial. This reservoir as planned would store 30 per cent more water than the Boulder Canyon Reservoir, the production of power would be much greater, and the cost would be less. However, on account of the radical design and proposed methods of construction, the project should be submitted to the best engineering talent in the world before it can be right or wise to adopt it.

WATER RIGHTS.

The Supreme Court of the United States has decided that in the case of interstate streams in the arid region, neither the riparian theory of water rights nor the priority of appropriation theory can obtain, but that each State is entitled to benefit from the river to substantial benefits. Presumably, the distribution of benefits must be made by the Federal court. But in the case of the Colorado River, where there is water enough for all, there seems to be no necessity for any litigation.

The States of the upper basin seem to fear that the construction of large reservoirs will serve automatically to appropriate the waters of the river for use in the lower basin, and that additional development of irrigation in the upper States will be prevented. Oft-repeated assertions of the United States Geological Survey and the United States Reclamation Service that the water supply is ample and adequate for all of the irrigable lands of both upper and lower basins have not served to allay the fear. Another cause of alarm in Colorado is the doubt as to whether that State will be allowed to divert 310,000 acre-feet of water per year from the Colorado basin, through tunnels at narrow places in the watershed, for use on the plains north and east of Denver, as is desired.

The upper States therefore are demanding a guaranty of unrestricted irrigation development in the upper basin before they will lend their support or consent to a Federal project in the canyon region. The lower-basin States are asking for an allotment of the water supply among the seven States.

The wisdom of a perpetual guaranty or of an allotment of the waters of the river is questionable. On no other river basin has either been attempted. It is not possible to foresee conditions a hundred years ahead, or even 30 years ahead. All irrigators who are putting the water to beneficial use should be protected, but in principle it may be exceedingly dangerous to reserve a valuable water supply for a project which may prove to be of doubtful feasibility. If an allotment of the water is attempted, most of the seven States will advance extravagant claims to water. Some of the States most involved have no adequate conception of the feasibility of their projects, and no just allotment can be made without thorough surveys of all proposed irrigation lands. It is unlikely that any allotment can be proposed which will not be held up in some legislature for many years, and meanwhile the ruin of the Imperial Valley may be accomplished.

There is no necessity for a distribution of the unused water rights at this time. If the act to appropriate money for a Colorado River project shall state as follows: "Provided, That nothing in this act shall be so construed as to affect in any way the rights to the use of the waters of the Colorado Basin of any State or any part of a State,” then the upper States can not be affected adversely by the project.

The average annual discharge of the river into the Gulf of California is 13,000,000 acre-feet. The projects of the upper basin are such that probably no more than 3,000,000 acre-feet of water additional can be consumed in those projects, and the balance of 10,000,000 acre-feet is more than twice as much as the States of the lower basin can use at least until a different economic order shall prevail.

Congress, through the Mondell Act, has provided for a Colorado River Commission consisting of one representative from each of the seven States and one from the Federal Government. The commission is now organized, with Herbert Hoover as its chairman, representing the Federal Government. The purpose expressed in the Mondell Act is the negotiation of a compact or agreement providing for an equitable division or apportionment of the water supply among the seven States.

NAVIGABILITY.

The existing treaty with Mexico declares the Colorado River to be a navigable stream, and a Federal court prohibited any action which might interfere with its navigability. The diversion of water for irrigation, therefore, is contrary to the treaty.